End-to-End Learning Framework for IMU-Based 6-DOF Odometry

This paper presents an end-to-end learning framework for performing 6-DOF odometry by using only inertial data obtained from a low-cost IMU. The proposed inertial odometry method allows leveraging inertial sensors that are widely available on mobile platforms for estimating their 3D trajectories. Fo...

Full description

Bibliographic Details
Main Authors: João Paulo Silva do Monte Lima, Hideaki Uchiyama, Rin-ichiro Taniguchi
Format: Article
Language:English
Published: MDPI AG 2019-08-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/19/17/3777
_version_ 1818035384942067712
author João Paulo Silva do Monte Lima
Hideaki Uchiyama
Rin-ichiro Taniguchi
author_facet João Paulo Silva do Monte Lima
Hideaki Uchiyama
Rin-ichiro Taniguchi
author_sort João Paulo Silva do Monte Lima
collection DOAJ
description This paper presents an end-to-end learning framework for performing 6-DOF odometry by using only inertial data obtained from a low-cost IMU. The proposed inertial odometry method allows leveraging inertial sensors that are widely available on mobile platforms for estimating their 3D trajectories. For this purpose, neural networks based on convolutional layers combined with a two-layer stacked bidirectional LSTM are explored from the following three aspects. First, two 6-DOF relative pose representations are investigated: one based on a vector in the spherical coordinate system, and the other based on both a translation vector and an unit quaternion. Second, the loss function in the network is designed with the combination of several 6-DOF pose distance metrics: mean squared error, translation mean absolute error, quaternion multiplicative error and quaternion inner product. Third, a multi-task learning framework is integrated to automatically balance the weights of multiple metrics. In the evaluation, qualitative and quantitative analyses were conducted with publicly-available inertial odometry datasets. The best combination of the relative pose representation and the loss function was the translation and quaternion together with the translation mean absolute error and quaternion multiplicative error, which obtained more accurate results with respect to state-of-the-art inertial odometry techniques.
first_indexed 2024-12-10T06:54:12Z
format Article
id doaj.art-b3b233d0b6fd499ebd7dc685cdda43df
institution Directory Open Access Journal
issn 1424-8220
language English
last_indexed 2024-12-10T06:54:12Z
publishDate 2019-08-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj.art-b3b233d0b6fd499ebd7dc685cdda43df2022-12-22T01:58:29ZengMDPI AGSensors1424-82202019-08-011917377710.3390/s19173777s19173777End-to-End Learning Framework for IMU-Based 6-DOF OdometryJoão Paulo Silva do Monte Lima0Hideaki Uchiyama1Rin-ichiro Taniguchi2Departamento de Computação, Universidade Federal Rural de Pernambuco, Recife 52171-900, BrazilLibrary, Kyushu University, Fukuoka 819-0395, JapanFaculty of Information Science and Electrical Engineering, Kyushu University, Fukuoka 819-0395, JapanThis paper presents an end-to-end learning framework for performing 6-DOF odometry by using only inertial data obtained from a low-cost IMU. The proposed inertial odometry method allows leveraging inertial sensors that are widely available on mobile platforms for estimating their 3D trajectories. For this purpose, neural networks based on convolutional layers combined with a two-layer stacked bidirectional LSTM are explored from the following three aspects. First, two 6-DOF relative pose representations are investigated: one based on a vector in the spherical coordinate system, and the other based on both a translation vector and an unit quaternion. Second, the loss function in the network is designed with the combination of several 6-DOF pose distance metrics: mean squared error, translation mean absolute error, quaternion multiplicative error and quaternion inner product. Third, a multi-task learning framework is integrated to automatically balance the weights of multiple metrics. In the evaluation, qualitative and quantitative analyses were conducted with publicly-available inertial odometry datasets. The best combination of the relative pose representation and the loss function was the translation and quaternion together with the translation mean absolute error and quaternion multiplicative error, which obtained more accurate results with respect to state-of-the-art inertial odometry techniques.https://www.mdpi.com/1424-8220/19/17/3777odometry6-DOFIMUneural networks
spellingShingle João Paulo Silva do Monte Lima
Hideaki Uchiyama
Rin-ichiro Taniguchi
End-to-End Learning Framework for IMU-Based 6-DOF Odometry
Sensors
odometry
6-DOF
IMU
neural networks
title End-to-End Learning Framework for IMU-Based 6-DOF Odometry
title_full End-to-End Learning Framework for IMU-Based 6-DOF Odometry
title_fullStr End-to-End Learning Framework for IMU-Based 6-DOF Odometry
title_full_unstemmed End-to-End Learning Framework for IMU-Based 6-DOF Odometry
title_short End-to-End Learning Framework for IMU-Based 6-DOF Odometry
title_sort end to end learning framework for imu based 6 dof odometry
topic odometry
6-DOF
IMU
neural networks
url https://www.mdpi.com/1424-8220/19/17/3777
work_keys_str_mv AT joaopaulosilvadomontelima endtoendlearningframeworkforimubased6dofodometry
AT hideakiuchiyama endtoendlearningframeworkforimubased6dofodometry
AT rinichirotaniguchi endtoendlearningframeworkforimubased6dofodometry