A Practical Representation of Time for the Human Behaviour Modelling

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F18%3A00328296" target="_blank" >RIV/68407700:21230/18:00328296 - isvavai.cz</a>

Výsledek na webu

<a href="http://www.ssds.sk/casopis/archiv/2018/fss0218.pdf#page=63" target="_blank" >http://www.ssds.sk/casopis/archiv/2018/fss0218.pdf#page=63</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

A Practical Representation of Time for the Human Behaviour Modelling

Popis výsledku v původním jazyce

This paper proposes a representation of the time domain intended for mobile robots which operate in human-populated environments.The method aims to identify and efficiently represent patterns of human habits, which are driven by periodic processes, such as the daily cycle.The core idea is to identify periodicities in the data observed by the robot and to project the time onto a series of circles, which represent the identified periodicities.This representation ensures that Euclidean distance of periodically-occurring events is low even if these events are temporally distant.This property allows to cluster events that occur at similar times of a day or similar days of a week etc. In the use-cases presented, we demonstrate that the method allows for temporally dependent anomaly detection and it can predictthefuture presence of people across large areas.The experiments indicate that the method detection reliability and prediction accuracy outperforms state-of-the-art tools used in statistical analysis for autonomous robots.

Název v anglickém jazyce

A Practical Representation of Time for the Human Behaviour Modelling

Popis výsledku anglicky

This paper proposes a representation of the time domain intended for mobile robots which operate in human-populated environments.The method aims to identify and efficiently represent patterns of human habits, which are driven by periodic processes, such as the daily cycle.The core idea is to identify periodicities in the data observed by the robot and to project the time onto a series of circles, which represent the identified periodicities.This representation ensures that Euclidean distance of periodically-occurring events is low even if these events are temporally distant.This property allows to cluster events that occur at similar times of a day or similar days of a week etc. In the use-cases presented, we demonstrate that the method allows for temporally dependent anomaly detection and it can predictthefuture presence of people across large areas.The experiments indicate that the method detection reliability and prediction accuracy outperforms state-of-the-art tools used in statistical analysis for autonomous robots.

Druh

J_ost - Ostatní články v recenzovaných periodicích

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/GJ17-27006Y" target="_blank" >GJ17-27006Y: Prostorově temporální representace pro dlouhodobou navigaci mobilních robotů</a><br>

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2018

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Forum Statisticum Slovacum

ISSN

1336-7420

e-ISSN

—

Svazek periodika

14

Číslo periodika v rámci svazku

2

Stát vydavatele periodika

SK - Slovenská republika

Počet stran výsledku

15

Strana od-do

61-75

Kód UT WoS článku

—

EID výsledku v databázi Scopus

—