Elliptical Model-Based Robust Pedestrian Pose Tracking by LiDAR via Bi-Level Moving Horizon Random Sample Consensus

Recent advancements in autonomous vehicle research have heightened the need for precise pedestrian pose tracking to ensure safe navigation. Utilizing LiDAR sensors, conventional bounding box (BB) methods are generally used but are sensitive to limb motion. The previous moving horizon-based maximum l...

Full description

Saved in:
Bibliographic Details
Main Authors: Haziq Muhammad, Zool H. Ismail, Kazuma Sekiguchi, Kenichiro Nonaka
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10971966/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850170083088793600
author Haziq Muhammad
Zool H. Ismail
Kazuma Sekiguchi
Kenichiro Nonaka
author_facet Haziq Muhammad
Zool H. Ismail
Kazuma Sekiguchi
Kenichiro Nonaka
author_sort Haziq Muhammad
collection DOAJ
description Recent advancements in autonomous vehicle research have heightened the need for precise pedestrian pose tracking to ensure safe navigation. Utilizing LiDAR sensors, conventional bounding box (BB) methods are generally used but are sensitive to limb motion. The previous moving horizon-based maximum likelihood estimation sample consensus (MHESAC) study addressed this robustness limitation but suffered from extensive point cloud (PC) sampling, limiting real-time applications. This paper proposes an elliptical pedestrian model tracking within a sequential bi-level MHESAC (B-MHESAC) framework that combines random sampling consensus (RANSAC) for local model fitting per time step and maximum likelihood estimation sample consensus (MLESAC) for optimal model sequence selection throughout the horizon, significantly reducing PC sampling while preserving accuracy. Under significant outlier contamination, B-MHESAC requires less trials than MHESAC for multiple horizons and samples. Moreover, at long horizon lengths, B-MHESAC required trial’s exponential growth is at least reduced to the fifth root of MHESAC, allowing real-time estimations. The proposed method was evaluated in an indoor single pedestrian tracking experiment. Experimental results demonstrate that B-MHESAC is more robust than conventional RANSAC in model fitting. Additionally, the elliptical model-based B-MHESAC’s tracking performance is competitive with the MHESAC method’s and superior to other Kalman filter-based BB and circle model methods.
format Article
id doaj-art-b581ee5c03cb45f5b088003d59a0e827
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-b581ee5c03cb45f5b088003d59a0e8272025-08-20T02:20:33ZengIEEEIEEE Access2169-35362025-01-0113728297284010.1109/ACCESS.2025.356303010971966Elliptical Model-Based Robust Pedestrian Pose Tracking by LiDAR via Bi-Level Moving Horizon Random Sample ConsensusHaziq Muhammad0https://orcid.org/0009-0006-2941-9501Zool H. Ismail1Kazuma Sekiguchi2Kenichiro Nonaka3https://orcid.org/0000-0003-0532-0495Advanced Control System Engineering Laboratory, Tokyo City University, Setagaya City, Tokyo, JapanAdvanced Control System Engineering Laboratory, Tokyo City University, Setagaya City, Tokyo, JapanAdvanced Control System Engineering Laboratory, Tokyo City University, Setagaya City, Tokyo, JapanAdvanced Control System Engineering Laboratory, Tokyo City University, Setagaya City, Tokyo, JapanRecent advancements in autonomous vehicle research have heightened the need for precise pedestrian pose tracking to ensure safe navigation. Utilizing LiDAR sensors, conventional bounding box (BB) methods are generally used but are sensitive to limb motion. The previous moving horizon-based maximum likelihood estimation sample consensus (MHESAC) study addressed this robustness limitation but suffered from extensive point cloud (PC) sampling, limiting real-time applications. This paper proposes an elliptical pedestrian model tracking within a sequential bi-level MHESAC (B-MHESAC) framework that combines random sampling consensus (RANSAC) for local model fitting per time step and maximum likelihood estimation sample consensus (MLESAC) for optimal model sequence selection throughout the horizon, significantly reducing PC sampling while preserving accuracy. Under significant outlier contamination, B-MHESAC requires less trials than MHESAC for multiple horizons and samples. Moreover, at long horizon lengths, B-MHESAC required trial’s exponential growth is at least reduced to the fifth root of MHESAC, allowing real-time estimations. The proposed method was evaluated in an indoor single pedestrian tracking experiment. Experimental results demonstrate that B-MHESAC is more robust than conventional RANSAC in model fitting. Additionally, the elliptical model-based B-MHESAC’s tracking performance is competitive with the MHESAC method’s and superior to other Kalman filter-based BB and circle model methods.https://ieeexplore.ieee.org/document/10971966/Light detection and rangingmaximum likelihood estimationmodel fittingmoving horizon estimationpedestrian trackingrandom sample consensus
spellingShingle Haziq Muhammad
Zool H. Ismail
Kazuma Sekiguchi
Kenichiro Nonaka
Elliptical Model-Based Robust Pedestrian Pose Tracking by LiDAR via Bi-Level Moving Horizon Random Sample Consensus
IEEE Access
Light detection and ranging
maximum likelihood estimation
model fitting
moving horizon estimation
pedestrian tracking
random sample consensus
title Elliptical Model-Based Robust Pedestrian Pose Tracking by LiDAR via Bi-Level Moving Horizon Random Sample Consensus
title_full Elliptical Model-Based Robust Pedestrian Pose Tracking by LiDAR via Bi-Level Moving Horizon Random Sample Consensus
title_fullStr Elliptical Model-Based Robust Pedestrian Pose Tracking by LiDAR via Bi-Level Moving Horizon Random Sample Consensus
title_full_unstemmed Elliptical Model-Based Robust Pedestrian Pose Tracking by LiDAR via Bi-Level Moving Horizon Random Sample Consensus
title_short Elliptical Model-Based Robust Pedestrian Pose Tracking by LiDAR via Bi-Level Moving Horizon Random Sample Consensus
title_sort elliptical model based robust pedestrian pose tracking by lidar via bi level moving horizon random sample consensus
topic Light detection and ranging
maximum likelihood estimation
model fitting
moving horizon estimation
pedestrian tracking
random sample consensus
url https://ieeexplore.ieee.org/document/10971966/
work_keys_str_mv AT haziqmuhammad ellipticalmodelbasedrobustpedestrianposetrackingbylidarviabilevelmovinghorizonrandomsampleconsensus
AT zoolhismail ellipticalmodelbasedrobustpedestrianposetrackingbylidarviabilevelmovinghorizonrandomsampleconsensus
AT kazumasekiguchi ellipticalmodelbasedrobustpedestrianposetrackingbylidarviabilevelmovinghorizonrandomsampleconsensus
AT kenichirononaka ellipticalmodelbasedrobustpedestrianposetrackingbylidarviabilevelmovinghorizonrandomsampleconsensus