Learning intentions for improved human motion prediction

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

16 Citaties (Scopus)

Uittreksel

For many tasks robots need to operate in human populated environments. Human motion prediction is gaining importance since this helps minimizing the hinder robots cause during the execution of these tasks. The concept of social forces defines virtual repelling and attracting forces from and to obstacles and points of interest. These social forces can be used to model typical human movements given an environment and a person’s intention. This work shows how such models can exploit typical motion patterns summarized by growing hidden Markov models (GHMMs) that can be learned from data online and without human intervention. An extensive series of experiments shows that exploiting a person’s intended position estimated using a GHMM within a social forces based motion model yields a significant performance gain in comparison with the standard constant velocity-based models.
TaalEngels
Pagina's591-602
Aantal pagina's12
TijdschriftRobotics and Autonomous Systems
Volume62
Nummer van het tijdschrift4
DOI's
StatusGepubliceerd - 2014

Vingerafdruk

Motion
Prediction
Hidden Markov models
Markov Model
Person
Robot
Robots
Model
Series
Learning
Human
Experiment
Experiments
Standards
Concepts
Movement

Citeer dit

@article{33192ff3d2c6431587f11cd26f75fb0a,
title = "Learning intentions for improved human motion prediction",
abstract = "For many tasks robots need to operate in human populated environments. Human motion prediction is gaining importance since this helps minimizing the hinder robots cause during the execution of these tasks. The concept of social forces defines virtual repelling and attracting forces from and to obstacles and points of interest. These social forces can be used to model typical human movements given an environment and a person’s intention. This work shows how such models can exploit typical motion patterns summarized by growing hidden Markov models (GHMMs) that can be learned from data online and without human intervention. An extensive series of experiments shows that exploiting a person’s intended position estimated using a GHMM within a social forces based motion model yields a significant performance gain in comparison with the standard constant velocity-based models.",
author = "J. Elfring and {Molengraft, van de}, M.J.G. and M. Steinbuch",
year = "2014",
doi = "10.1016/j.robot.2014.01.003",
language = "English",
volume = "62",
pages = "591--602",
journal = "Robotics and Autonomous Systems",
issn = "0921-8890",
publisher = "Elsevier",
number = "4",

}

Learning intentions for improved human motion prediction. / Elfring, J.; Molengraft, van de, M.J.G.; Steinbuch, M.

In: Robotics and Autonomous Systems, Vol. 62, Nr. 4, 2014, blz. 591-602.

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

TY - JOUR

T1 - Learning intentions for improved human motion prediction

AU - Elfring,J.

AU - Molengraft, van de,M.J.G.

AU - Steinbuch,M.

PY - 2014

Y1 - 2014

N2 - For many tasks robots need to operate in human populated environments. Human motion prediction is gaining importance since this helps minimizing the hinder robots cause during the execution of these tasks. The concept of social forces defines virtual repelling and attracting forces from and to obstacles and points of interest. These social forces can be used to model typical human movements given an environment and a person’s intention. This work shows how such models can exploit typical motion patterns summarized by growing hidden Markov models (GHMMs) that can be learned from data online and without human intervention. An extensive series of experiments shows that exploiting a person’s intended position estimated using a GHMM within a social forces based motion model yields a significant performance gain in comparison with the standard constant velocity-based models.

AB - For many tasks robots need to operate in human populated environments. Human motion prediction is gaining importance since this helps minimizing the hinder robots cause during the execution of these tasks. The concept of social forces defines virtual repelling and attracting forces from and to obstacles and points of interest. These social forces can be used to model typical human movements given an environment and a person’s intention. This work shows how such models can exploit typical motion patterns summarized by growing hidden Markov models (GHMMs) that can be learned from data online and without human intervention. An extensive series of experiments shows that exploiting a person’s intended position estimated using a GHMM within a social forces based motion model yields a significant performance gain in comparison with the standard constant velocity-based models.

U2 - 10.1016/j.robot.2014.01.003

DO - 10.1016/j.robot.2014.01.003

M3 - Article

VL - 62

SP - 591

EP - 602

JO - Robotics and Autonomous Systems

T2 - Robotics and Autonomous Systems

JF - Robotics and Autonomous Systems

SN - 0921-8890

IS - 4

ER -