A Bidirectional LSTM for Estimating Dynamic Human Velocities from a Single IMU

The main challenge in estimating human velocity from noisy Inertial Measurement Units (IMUs) are the errors that accumulate by integrating noisy accelerometer signals over a long time. Known approaches that work on step length estimation are optimized for a specific application, sensor position, and movement type, require an exhaustive (manual) parameter tuning, and can thus not be applied to other movement types or to a broader range of applications. Moreover, varying dynamics (as they are present for instance in sports applications) cause abrupt and unpredictable changes in step frequency or step length and hence result in erroneous velocity estimates.We use machine learning (ML) and deep learning (DL) to estimate a human’s velocity. Our approach is robust to varying motion states and orientation changes in dynamic situations. On data from a single un-calibrated IMU, our novel recurrent model not only outperforms the state-of-the-art on instantaneous velocity (≤0.10 m/s) and on traveled distance (≤29 m/km). It can also generalize to different and varying rates of motion and provides accurate and precise velocity estimates.

• Publication year

  2019

• Venue

  International Conference on Indoor Positioning and Indoor Navigation

• Publication date

  2019-09-01

• Fields of study

  Computer Science, Engineering

• Identifiers
  DOI 10.1109/IPIN.2019.8911814
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Pedestrian Stride-Length Estimation Based on LSTM and Denoising Autoencoders

  2019cited by this paper
• A Deep Learning Approach to Position Estimation from Channel Impulse Responses †

  2019cited by this paper
• Error Analysis of PDR System Using Dual Foot-mounted IMU

  2019influential reference
• Accurate Step Length Estimation for Pedestrian Dead Reckoning Localization Using Stacked Autoencoders

  2019influential reference
• Comparison of Step Length Estimation Models Using Inertial Sensor on Pelvis

  2019cited by this paper
• Zero-Velocity Detection—A Bayesian Approach to Adaptive Thresholding

  2019cited by this paper
• OxIOD: The Dataset for Deep Inertial Odometry

  2018cited by this paper
• A MEMS IMU De-Noising Method Using Long Short Term Memory Recurrent Neural Networks (LSTM-RNN)

  2018influential reference
• 1D CNN with BLSTM for automated classification of fixations, saccades, and smooth pursuits

  2018cited by this paper
• Error Analysis of ZUPT-Aided Pedestrian Inertial Navigation

  2018influential reference
• Motion Context Adaptive Fusion of Inertial and Visual Pedestrian Navigation

  2018cited by this paper
• IONet: Learning to Cure the Curse of Drift in Inertial Odometry

  2018cited by this paper
• LSTM-Based Zero-Velocity Detection for Robust Inertial Navigation

  2018influential reference
• DePedo: Anti Periodic Negative-Step Movement Pedometer with Deep Convolutional Neural Networks

  2018cited by this paper
• Step Length Estimation Methods Based on Inertial Sensors: A Review

  2018cited by this paper
• Supervised Learning for Yaw Orientation Estimation

  2018influential reference
• Recurrent Neural Networks on Drifting Time-of-Flight Measurements

  2018influential reference
• Comparison of step length and heading estimation methods for indoor environments

  2017cited by this paper
• An indoor positioning system using pedestrian dead reckoning with WiFi and map-matching aided

  2017cited by this paper
• Foot-mounted pedestrian navigation reference with tightly coupled GNSS carrier phases, inertial and magnetic data

  2017cited by this paper
• Simultaneous localization and mapping for pedestrians using low-cost ultra-wideband system and gyroscope

  2017influential reference
• RIDI: Robust IMU Double Integration

  2017cited by this paper
• Pattern Recognition And Machine Learning

  2016influential reference
• A Compositional Object-Based Approach to Learning Physical Dynamics

  2016cited by this paper
• Label-Free Supervision of Neural Networks with Physics and Domain Knowledge

  2016cited by this paper
• Regression Model-Based Walking Speed Estimation Using Wrist-Worn Inertial Sensor

  2016cited by this paper
• Combining CNN and BLSTM to Extract Textual and Acoustic Features for Recognizing Stances in Mandarin Ideological Debate Competition

  2016cited by this paper
• Walking compass with head-mounted IMU sensor

  2016influential reference
• A real-time indoor tracking system by fusing inertial sensor, radio signal and floor plan

  2016cited by this paper
• IMU/magnetometer based 3D indoor positioning for wheeled platforms in NLoS scenarios

  2016cited by this paper
• An advanced method for pedestrian dead reckoning using BLSTM-RNNs

  2015influential reference
• A realistic evaluation and comparison of indoor location technologies: experiences and lessons learned

  2015cited by this paper
• Comparison of misalignment estimation techniques between handheld device and walking directions

  2015cited by this paper
• A Convolutional Neural Network for Modelling Sentences

  2014influential reference
• Toward free-living walking speed estimation using Gaussian Process-based Regression with on-body accelerometers and gyroscopes

  2010cited by this paper
• Quaternion-based extended Kalman filter for determining orientation by inertial and magnetic sensing

  2006cited by this paper

• Measurement Uncertainty and Traceability in Upper Limb Rehabilitation Robotics: A Metrology-Oriented Review

  2026cites this paper
• Enhanced LSTM-attention model for high precision LiDAR/IMU tightly integrated navigation with velocity constraint

  2025cites this paper
• AttentionAR: AR Adaptation and Warning for Real-World Safety via Attention Modeling and MLLM Reasoning

  2025cites this paper
• UAV inertial navigation using deep neural networks in simulated environments

  2025cites this paper
• Tartan IMU: A Light Foundation Model for Inertial Positioning in Robotics

  2025cites this paper
• PDRNN: Modular Data-driven Pedestrian Dead Reckoning on Loosely Coupled Radio- and Inertial-Signalstreams

  2025cites this paper
• Reliability of a Low-Cost Inertial Measurement Unit (IMU) to Measure Punch and Kick Velocity

  2025cites this paper
• Constraints on the Retrieval 3-D Velocities by Hybrid Regression Network for GNSS/INS Positioning During GNSS Outages

  2025cites this paper
• Step Length and Gait Speed Estimation Using a Hearing Aid Integrated Accelerometer: A Comparison of Different Algorithms

  2024cites this paper
• Toward Persistent Spatial Awareness: A Review of Pedestrian Dead Reckoning-Centric Indoor Positioning With Smartphones

  2024cites this paper
• Reliable Heading Tracking for Pedestrian Road Crossing Prediction Using Commodity Devices

  2024cites this paper
• Analysis of Output Domain More Reliable for Deep Learning-Based Step Length Estimation

  2024cites this paper
• SdoNet: Speed Odometry Network and Noise Adapter for Vehicle Integrated Navigation

  2023cites this paper
• Deep Learning for Inertial Positioning: A Survey

  2023cites this paper
• Factor Graph Framework for Smartphone Indoor Localization: Integrating Data-Driven PDR and Wi-Fi RTT/RSS Ranging

  2023cites this paper
• Non-holonomic constraint (NHC)-assisted GNSS/SINS positioning using a vehicle motion state classification (VMSC)-based convolution neural network

  2023cites this paper
• Real-Time Speed Monitoring of Elevator System Based on Low-Cost IMU

  2023cites this paper
• Pedestrian Inertial Navigation with Multi-Head CNN

  2023cites this paper
• Navigation magnéto-inertielle - Filtrage de Kalman et intelligence artificielle

  2023cites this paper
• Pedestrian Inertial Navigation with Multi-Head LSTM Including Attention

  2023cites this paper
• A Deep Learning Approach for Biped Robot Locomotion Interface Using a Single Inertial Sensor

  2023cites this paper
• Walking Direction Estimation Using Smartphone Sensors: A Deep Network-Based Framework

  2022cites this paper
• Heading Estimation Based on Magnetometer Measurement using LSTM

  2022cites this paper
• A Survey on Artificial Intelligence for Pedestrian Navigation with Wearable Inertial Sensors

  2022cites this paper
• A feature fusion sequence learning approach for quantitative analysis of tremor symptoms based on digital handwriting

  2022cites this paper
• Survey of Machine Learning Methods Applied to Urban Mobility

  2022cites this paper
• Smartphone-Based Inertial Odometry for Blind Walkers

  2021cites this paper
• Estimating TOA Reliability With Variational Autoencoders

  2021cites this paper
• OdoNet: Untethered Speed Aiding for Vehicle Navigation Without Hardware Wheeled Odometer

  2021cites this paper
• Online Dynamic Window (ODW) Assisted Two-Stage LSTM Frameworks For Indoor Localization

  2021cites this paper
• Online Dynamic Window (ODW) Assisted 2-Stage LSTM Indoor Localization for Smart Phones

  2021cites this paper
• PDRNet: A Deep-Learning Pedestrian Dead Reckoning Framework

  2021cites this paper
• Machine Learning Driven Method for Indoor Positioning Using Inertial Measurement Unit

  2020cites this paper
• Machine Learning Improvements to Human Motion Tracking with IMUs

  2020cites this paper
• Machine Learning-Based Zero-Velocity Detection for Inertial Pedestrian Navigation

  2020cites this paper
• A Novel Adaptive Zero-Velocity Detector for Inertial Pedestrian Navigation Based on Optimal Interval Estimation

  2020cites this paper
• RNN-Aided Human Velocity Estimation from a Single IMU †

  2020influential citation
• NLOS Detection using UWB Channel Impulse Responses and Convolutional Neural Networks

  2020cites this paper
• StepNet—Deep Learning Approaches for Step Length Estimation

  2020cites this paper
• UWB Channel Impulse Responses for Positioning in Complex Environments: A Detailed Feature Analysis

  2019cites this paper
• Multiple Model-based Indoor Localization via Bluetooth Low Energy and Inertial Measurement Unit Sensors

  year unknowncites this paper