{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,16]],"date-time":"2026-04-16T06:34:36Z","timestamp":1776321276058,"version":"3.50.1"},"reference-count":53,"publisher":"MDPI AG","issue":"6","license":[{"start":{"date-parts":[[2023,3,17]],"date-time":"2023-03-17T00:00:00Z","timestamp":1679011200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Computational Infrastructure (NCI Australia)","award":["LE160100051"],"award-info":[{"award-number":["LE160100051"]}]},{"name":"Australian Government","award":["LE160100051"],"award-info":[{"award-number":["LE160100051"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>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 \u22640.4594 m, highlighting the effectiveness of our learning process used in the development of the models.<\/jats:p>","DOI":"10.3390\/s23063217","type":"journal-article","created":{"date-parts":[[2023,3,17]],"date-time":"2023-03-17T05:06:01Z","timestamp":1679029561000},"page":"3217","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":12,"title":["RIOT: Recursive Inertial Odometry Transformer for Localisation from Low-Cost IMU Measurements"],"prefix":"10.3390","volume":"23","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7912-8741","authenticated-orcid":false,"given":"James","family":"Brotchie","sequence":"first","affiliation":[{"name":"School of Science, RMIT University, Melbourne, VIC 3001, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8926-5539","authenticated-orcid":false,"given":"Wenchao","family":"Li","sequence":"additional","affiliation":[{"name":"School of Science, RMIT University, Melbourne, VIC 3001, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3505-9163","authenticated-orcid":false,"given":"Andrew D.","family":"Greentree","sequence":"additional","affiliation":[{"name":"ARC Centre of Excellence for Nanoscale BioPhotonics, School of Science, RMIT University, Melbourne, VIC 3001, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Allison","family":"Kealy","sequence":"additional","affiliation":[{"name":"School of Science, RMIT University, Melbourne, VIC 3001, Australia"},{"name":"Victorian Department of Environment, Land, Water and Planning, Melbourne, VIC 3000, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2023,3,17]]},"reference":[{"key":"ref_1","first-page":"140","article-title":"Analysis and modeling of inertial sensors using Allan variance","volume":"57","author":"Hou","year":"2007","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"319","DOI":"10.1080\/00207179208934317","article-title":"Neural networks for nonlinear dynamic system modelling and identification","volume":"56","author":"Chen","year":"1992","journal-title":"Int. J. Control."},{"key":"ref_3","unstructured":"Pascanu, R., Mikolov, T., and Bengio, Y. (2013, January 17\u201319). On the difficulty of training recurrent neural networks. Proceedings of the International Conference on Machine Learning, PMLR, Atlanta, GE, USA."},{"key":"ref_4","unstructured":"Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, \u0141., and Polosukhin, I. (2017, January 4\u20139). Attention is all you need. Proceedings of the Advances in Neural Information Processing Systems, Long Beach, CA, USA."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3505244","article-title":"Transformers in vision: A survey","volume":"54","author":"Khan","year":"2022","journal-title":"ACM Comput. Surv. (CSUR)"},{"key":"ref_6","unstructured":"Devlin, J., Chang, M.W., Lee, K., and Toutanova, K. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv."},{"key":"ref_7","first-page":"1","article-title":"A transformer self-attention model for time series forecasting","volume":"9","author":"Pazouki","year":"2021","journal-title":"J. Electr. Comput. Eng. Innov. (JECEI)"},{"key":"ref_8","unstructured":"Parmar, N., Vaswani, A., Uszkoreit, J., Kaiser, L., Shazeer, N., Ku, A., and Tran, D. (2018, January 10\u201315). Image transformer. Proceedings of the International Conference on Machine Learning. PMLR, Stockholm, Sweden."},{"key":"ref_9","unstructured":"Liu, P.J., Saleh, M., Pot, E., Goodrich, B., Sepassi, R., Kaiser, L., and Shazeer, N. (2018). Generating wikipedia by summarizing long sequences. arXiv."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Povey, D., Hadian, H., Ghahremani, P., Li, K., and Khudanpur, S. (2018, January 15\u201320). A time-restricted self-attention layer for ASR. Proceedings of the 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Calgary, AB, Canada.","DOI":"10.1109\/ICASSP.2018.8462497"},{"key":"ref_11","unstructured":"Huang, C.Z.A., Vaswani, A., Uszkoreit, J., Shazeer, N., Simon, I., Hawthorne, C., Dai, A.M., Hoffman, M.D., Dinculescu, M., and Eck, D. (2018). Music transformer. arXiv."},{"key":"ref_12","unstructured":"Merkx, D., and Frank, S.L. (2020). Comparing Transformers and RNNs on predicting human sentence processing data. arXiv."},{"key":"ref_13","unstructured":"Radford, A., Narasimhan, K., Salimans, T., and Sutskever, I. (2022, November 10). Improving Language Understanding by Generative Pre-Training. Available online: https:\/\/www.cs.ubc.ca\/~amuham01\/LING530\/papers\/radford2018improving.pdf."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Mahdi, A.E., Azouz, A., Abdalla, A.E., and Abosekeen, A. (2022). A machine learning approach for an improved inertial navigation system solution. Sensors, 22.","DOI":"10.3390\/s22041687"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Han, S., Meng, Z., Zhang, X., and Yan, Y. (2021). Hybrid deep recurrent neural networks for noise reduction of MEMS-IMU with static and dynamic conditions. Micromachines, 12.","DOI":"10.3390\/mi12020214"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Chen, H., Aggarwal, P., Taha, T.M., and Chodavarapu, V.P. (2018, January 23\u201326). Improving inertial sensor by reducing errors using deep learning methodology. Proceedings of the NAECON 2018-IEEE National Aerospace and Electronics Conference, Dayton, OH, USA.","DOI":"10.1109\/NAECON.2018.8556718"},{"key":"ref_17","first-page":"1","article-title":"A MEMS IMU gyroscope calibration method based on deep learning","volume":"71","author":"Huang","year":"2022","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Yan, H., Shan, Q., and Furukawa, Y. (2018, January 8\u201314). RIDI: Robust IMU double integration. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany.","DOI":"10.1007\/978-3-030-01261-8_38"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"4932","DOI":"10.1109\/JSEN.2021.3066840","article-title":"PDRNet: A deep-learning pedestrian dead reckoning framework","volume":"22","author":"Asraf","year":"2021","journal-title":"IEEE Sens. J."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Chen, C., Lu, X., Markham, A., and Trigoni, N. (2018, January 2\u20137). Ionet: Learning to cure the curse of drift in inertial odometry. Proceedings of the AAAI Conference on Artificial Intelligence, New Orleans, LA, USA.","DOI":"10.1609\/aaai.v32i1.12102"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Yan, H., Herath, S., and Furukawa, Y. (2019). Ronin: Robust neural inertial navigation in the wild: Benchmark, evaluations, and new methods. arXiv.","DOI":"10.1109\/ICRA40945.2020.9196860"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27\u201330). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_23","unstructured":"Khorrambakht, R., Lu, C.X., Damirchi, H., Chen, Z., and Li, Z. (2021). Deep Inertial Odometry with Accurate IMU Preintegration. arXiv."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"5653","DOI":"10.1109\/LRA.2020.3007421","article-title":"Tlio: Tight learned inertial odometry","volume":"5","author":"Liu","year":"2020","journal-title":"IEEE Robot. Autom. Lett."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"168806","DOI":"10.1109\/ACCESS.2021.3135012","article-title":"Evaluating Tracking Rotations Using Maximal Entropy Distributions for Smartphone Applications","volume":"9","author":"Brotchie","year":"2021","journal-title":"IEEE Access"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Sun, S., Melamed, D., and Kitani, K. (2021, January 2\u20139). IDOL: Inertial deep orientation-estimation and localization. Proceedings of the AAAI Conference on Artificial Intelligence, Virtually.","DOI":"10.1609\/aaai.v35i7.16763"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TIM.2021.3123218","article-title":"Pose-invariant inertial odometry for pedestrian localization","volume":"70","author":"Wang","year":"2021","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Cao, X., Zhou, C., Zeng, D., and Wang, Y. (2022, January 18\u201324). RIO: Rotation-equivariance supervised learning of robust inertial odometry. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.","DOI":"10.1109\/CVPR52688.2022.00650"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Wang, Y., Cheng, H., and Meng, M.Q.H. (2022, January 23\u201327). A2DIO: Attention-Driven Deep Inertial Odometry for Pedestrian Localization based on 6D IMU. Proceedings of the 2022 International Conference on Robotics and Automation (ICRA), Philadelphia, PA, USA.","DOI":"10.1109\/ICRA46639.2022.9811714"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"e00938","DOI":"10.1016\/j.heliyon.2018.e00938","article-title":"State-of-the-art in artificial neural network applications: A survey","volume":"4","author":"Abiodun","year":"2018","journal-title":"Heliyon"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"3370","DOI":"10.1109\/TNNLS.2019.2891257","article-title":"Multistep prediction of dynamic systems with recurrent neural networks","volume":"30","author":"Mohajerin","year":"2019","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1186\/s40537-021-00444-8","article-title":"Review of deep learning: Concepts, CNN architectures, challenges, applications, future directions","volume":"8","author":"Alzubaidi","year":"2021","journal-title":"J. Big Data"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Jim\u00e9nez, A.R., Seco, F., Prieto, J.C., and Guevara, J. (2010, January 11\u201312). Indoor pedestrian navigation using an INS\/EKF framework for yaw drift reduction and a foot-mounted IMU. Proceedings of the 2010 7th Workshop on Positioning, Navigation and Communication, Dresden, Germany.","DOI":"10.1109\/WPNC.2010.5649300"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Brotchie, J., Shao, W., Li, W., and Kealy, A. (2022). Leveraging Self-Attention Mechanism for Attitude Estimation in Smartphones. Sensors, 22.","DOI":"10.3390\/s22229011"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"12","DOI":"10.2514\/1.22452","article-title":"Survey of nonlinear attitude estimation methods","volume":"30","author":"Crassidis","year":"2007","journal-title":"J. Guid. Control Dyn."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Coviello, G., Avitabile, G., Florio, A., and Talarico, C. (2020, January 9\u201312). A study on IMU sampling rate mismatch for a wireless synchronized platform. Proceedings of the 2020 IEEE 63rd International Midwest Symposium on Circuits and Systems (MWSCAS), Springfield, MA, USA.","DOI":"10.1109\/MWSCAS48704.2020.9184531"},{"key":"ref_37","unstructured":"Maas, A.L., Hannun, A.Y., and Ng, A.Y. (2013, January 16\u201321). Rectifier nonlinearities improve neural network acoustic models. Proceedings of the ICML Citeseer, Atlanta, GA, USA."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Kok, M., Hol, J.D., and Sch\u00f6n, T.B. (2017). Using inertial sensors for position and orientation estimation. arXiv.","DOI":"10.1561\/9781680833577"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"155","DOI":"10.1007\/s10851-009-0161-2","article-title":"Metrics for 3D rotations: Comparison and analysis","volume":"35","author":"Huynh","year":"2009","journal-title":"J. Math. Imaging Vis."},{"key":"ref_40","unstructured":"Kingma, D.P., and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"376","DOI":"10.1109\/34.88573","article-title":"Least-squares estimation of transformation parameters between two point patterns","volume":"13","author":"Umeyama","year":"1991","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Yang, S., Yu, X., and Zhou, Y. (2020, January 12\u201314). Lstm and gru neural network performance comparison study: Taking yelp review dataset as an example. Proceedings of the 2020 International Workshop on Electronic Communication and Artificial Intelligence (IWECAI), Shanghai, China.","DOI":"10.1109\/IWECAI50956.2020.00027"},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"40","DOI":"10.3389\/frai.2020.00040","article-title":"Are GRU cells more specific and LSTM cells more sensitive in motive classification of text?","volume":"3","author":"Gruber","year":"2020","journal-title":"Front. Artif. Intell."},{"key":"ref_44","unstructured":"Cahuantzi, R., Chen, X., and G\u00fcttel, S. (2021). A comparison of LSTM and GRU networks for learning symbolic sequences. arXiv."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"6085","DOI":"10.1038\/s41598-018-24271-9","article-title":"Recurrent neural networks for multivariate time series with missing values","volume":"8","author":"Che","year":"2018","journal-title":"Sci. Rep."},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"444","DOI":"10.3390\/ai2030028","article-title":"RIANN\u2014A robust neural network outperforms attitude estimation filters","volume":"2","author":"Weber","year":"2021","journal-title":"AI"},{"key":"ref_47","unstructured":"Chen, C., Zhao, P., Lu, C.X., Wang, W., Markham, A., and Trigoni, N. (2018). Oxiod: The dataset for deep inertial odometry. arXiv."},{"key":"ref_48","doi-asserted-by":"crossref","unstructured":"Vleugels, R., Van Herbruggen, B., Fontaine, J., and De Poorter, E. (2021). Ultra-Wideband Indoor Positioning and IMU-Based Activity Recognition for Ice Hockey Analytics. Sensors, 21.","DOI":"10.3390\/s21144650"},{"key":"ref_49","doi-asserted-by":"crossref","first-page":"8617","DOI":"10.1109\/TVT.2021.3101515","article-title":"Asynchronous Sensor Fusion of GPS, IMU and CAN-Based Odometry for Heavy-Duty Vehicles","volume":"70","author":"Armesto","year":"2021","journal-title":"IEEE Trans. Veh. Technol."},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"1467","DOI":"10.1109\/TII.2022.3158935","article-title":"A Function Approximator Model for Robust Online Foot Angle Trajectory Prediction Using a Single IMU Sensor: Implication for Controlling Active Prosthetic Feet","volume":"19","author":"Dey","year":"2022","journal-title":"IEEE Trans. Ind. Inform."},{"key":"ref_51","unstructured":"Vicon (2021, November 01). Vicon Motion Capture Systems. Available online: https:\/\/www.vicon.com\/?s=Motion%20Capture%20Systems."},{"key":"ref_52","unstructured":"Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep Learning, MIT Press."},{"key":"ref_53","doi-asserted-by":"crossref","first-page":"2224","DOI":"10.1109\/JSEN.2017.2660522","article-title":"Improved Wi-Fi RSSI Measurement for Indoor Localization","volume":"17","author":"Xue","year":"2017","journal-title":"IEEE Sens. J."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/23\/6\/3217\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T18:57:31Z","timestamp":1760122651000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/23\/6\/3217"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,3,17]]},"references-count":53,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2023,3]]}},"alternative-id":["s23063217"],"URL":"https:\/\/doi.org\/10.3390\/s23063217","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,3,17]]}}}