his.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Signals for Active Safety Systems to Detect Cyclists and Their Intentions in Traffic
University of Skövde, School of Informatics. University of Skövde, The Informatics Research Centre. (Interaction Lab (ILAB))ORCID iD: 0000-0002-1227-6843
2018 (English)Conference paper, Oral presentation with published abstract (Refereed)
Abstract [en]

Objectives: Human cognition is importantly predictive. This predictive ability can also be applied to predict the future actions of cyclists in traffic. Active safety systems in (semi-)autonomous vehicles will likely need to detect and predict human actions occurring in different traffic situations.

Results from two experiments demonstrate the effect of different patterns of human movement on predicting the behavior of the cyclists and the distance it takes drivers to detect the cyclists in a city environment. This research was carried out by observing recorded sequences on a computer but also in a driving simulator in order to include more naturalistic conditions and to achieve a high level of experimental control. As a complement to our previous research (Hemeren et al., 2014), we aimed to determine the distance at which drivers would detect and predict cyclists’ behavior.

Methods: Participants in both experiments (90 participants in experiment 1 and 24 in experiment 2) observed video-recorded cyclists wearing three different patterns of reflective clothing (Fig. 1): biomotion, vest and the legal minimum requirement (legal), in which no reflector material was worn by the cyclists. In experiment 1, participants were instructed to predict if an approaching cyclist would make a left-turn or continue straight on when approaching a crossing. This task was also performed during daylight, dusk and at night. In the second experiment, participants in a driving simulator indicated (as a secondary task) when they saw a cyclist riding along the side of the road.

Results: The biomotion reflective clothing led to a prediction accuracy of 88% for cyclists’ intentions at 9 meters before a crossing for the nighttime condition. For the legal minimum, the result was 59% and for the vest 67%. Detection distance in the driving simulator was also significantly greater for the biomotion condition compared to the legal and vest conditions. Visual detection is almost twice the distance for biomotion compared to the other two reflective clothing conditions.

Conclusions: The results point to the critical role that biological motion can play on predicting the intention and detection of cyclists in traffic. This information can be used to inform (semi-)autonomous systems of human intentions in traffic.

Place, publisher, year, edition, pages
2018.
National Category
Interaction Technologies
Research subject
Interaction Lab (ILAB)
Identifiers
URN: urn:nbn:se:his:diva-16358OAI: oai:DiVA.org:his-16358DiVA, id: diva2:1260519
Conference
The Eye, The Brain & The Auto - 8 th World Research Congress on Vision and Driving, Detroit, USA, October 7-9, 2018
Funder
Swedish Transport Administration, TRV 2015/85218Available from: 2018-11-03 Created: 2018-11-03 Last updated: 2019-05-23Bibliographically approved

Open Access in DiVA

fulltext(99 kB)5 downloads
File information
File name FULLTEXT01.pdfFile size 99 kBChecksum SHA-512
da42ee3f4dd08947f49d4deec5d4cdc303d18ea34fcfeddfa6bf78fe54109015651f7e6a512d777c06e9ecd5d9876e4df2b5d77ba3fb9c4fab8457e61eac9a0c
Type fulltextMimetype application/pdf

Authority records BETA

Hemeren, Paul

Search in DiVA

By author/editor
Hemeren, Paul
By organisation
School of InformaticsThe Informatics Research Centre
Interaction Technologies

Search outside of DiVA

GoogleGoogle Scholar
Total: 5 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 449 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf