Deep Learning based Pedestrian Inertial Navigation:
Methods and Dataset

Changhao Chen
Chris Xiaoxuan Lu
Andrew Markham
Niki Trigoni
Department of Computer Science, University of Oxford



Modern inertial measurements units (IMUs) are small, cheap, energy efficient, and widely employed in smart devices and mobile robots. Exploiting inertial data for accurate and reliable pedestrian navigation supports is a key component for emerging Internet-of-Things applications and services. %has attracted significant research and industrial interests, as IMU measurements are completely ego-centric and generally environment agnostic. %Recent studies showed that the notorious issue of drift can be significantly alleviated by using deep neural networks (DNNs). Recently, there has been a growing interest in applying deep neural networks (DNNs) to motion sensing and location estimation. However, the lack of sufficient labelled data for training and evaluating architecture benchmarks has limited the adoption of DNNs in IMU-based tasks. We present and release the Oxford Inertial Odometry Dataset (OxIOD), a first-of-its-kind public dataset for deep learning based inertial navigation research, with fine-grained ground-truth on all sequences. Furthermore, to enable more efficient inference at the edge, we propose a novel lightweight framework to learn and reconstruct pedestrian trajectories from raw IMU data. Extensive experiments show the effectiveness of our dataset and methods in achieving accurate data-driven pedestrian inertial navigation on resource-constrained devices. With the release of this large-scale diverse dataset, it is our hope that it will prove valuable to the community and enable future research in long-term ubiquitous ego-motion estimation.


Oxford Inertial Odometry Dataset


[Dataset (1.01G)]






Papers

Deep Learning based Pedestrian Inertial Navigation: Methods, Dataset and On-Device Inference

Changhao Chen, Peijun Zhao, Chris Xiaoxuan Lu, Wei Wang, Andrew Markham, Niki Trigoni

In IEEE Internet of Things Journal [PDF] [Bibtex]
OxIOD: The Dataset for Deep Inertial Odometry

Changhao Chen, Peijun Zhao, Chris Xiaoxuan Lu, Wei Wang, Andrew Markham, Niki Trigoni

Technical Report. arXiv:1809.07491 [PDF (2.4 MB)] [Bibtex]
MotionTransformer: Transferring Neural Inertial Tracking Between Domains

Changhao Chen, Yishu Miao, Chris Xiaoxuan Lu, Linhai Xie, Phil Blunsom, Andrew Markham, Niki Trigoni

In AAAI 2019 [PDF (1.3 MB)] [Bibtex]
Deep Neural Network Based Inertial Odometry Using Low-cost Inertial Measurement Units

Changhao Chen, Chris Xiaoxuan Lu, Johan Wahlstrom, Andrew Markham, Niki Trigoni

In IEEE Transactions on Mobile Computing [PDF] [Bibtex]
IONet: Learning to Cure the Curse of Drift in Inertial Odometry

Changhao Chen, Chris Xiaoxuan Lu, Andrew Markham, Niki Trigoni

In AAAI 2018 [PDF (2.4 MB)] [Bibtex]


Supplemental video