Road user detection with convolutional neural networks: An application to the autonomous shuttle WEpod

Floris Gaisser, Pieter Jonker

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

8 Citations (Scopus)

Abstract

Over a million fatal accidents occur every year with road vehicles. Road user detection for Advanced Driver Assistance Systems and Autonomous Vehicles could significantly reduce the number of accidents. Despite the research focus on road user detection and such systems, there is a surprising lack of research in real-world applications. In this work, radar and camera data are combined on an autonomous shuttle called `WEpod', driving on the public road in Wageningen, The Netherlands. With experiments we show that our method reduces the candidate region margin to 0.2m and reduces the miss rate significantly. Furthermore, our specifically trained Convolutional Neural Network improves the performance by 1.4% over vision-based road user detection, and combined with radars we improve by 7.6%. Finally, with our approach we show a performance of 95.1% on the WEpod while driving on the public road.
Original languageEnglish
Title of host publicationProceedings of the 15th IAPR International Conference on Machine Vision Applications
Subtitle of host publication- MVA2017
EditorsHiroshi Ishikawa, Norimichi Ukita, Hitoshi Habe
Place of PublicationPiscataway, NJ, USA
PublisherIEEE
Pages101-104
ISBN (Print)978-4-9011-2216-0
DOIs
Publication statusPublished - 2017
EventMVA2017: 15th IAPR International Conference on Machine Vision Applications - Nagoya, Japan
Duration: 8 May 201712 May 2017

Conference

ConferenceMVA2017: 15th IAPR International Conference on Machine Vision Applications
CountryJapan
CityNagoya
Period8/05/1712/05/17

Keywords

  • Roads
  • Radar Detection
  • Vehicle dynamics
  • Cameras
  • Sensors
  • Visualization

Fingerprint Dive into the research topics of 'Road user detection with convolutional neural networks: An application to the autonomous shuttle WEpod'. Together they form a unique fingerprint.

Cite this