Dynamic representations for autonomous driving

J.S. Olier Jauregui, P. Marin, D. Martin (Supervisor), L. Marcenaro (Supervisor), E.I. Barakova (Supervisor), G.W.M. Rauterberg (Supervisor), C. Regazzoni (Supervisor)

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

4 Citations (Scopus)
3 Downloads (Pure)

Abstract

This paper presents a method for observational learning in autonomous agents. A formalism based on deep learning implementations of variational methods and Bayesian filtering theory is presented. It is explained how the proposed method is capable of modeling the environment to mimic behaviors in an observed interaction by building internal representations and discovering temporal and causal relations. The method is evaluated in a typical surveillance scenario, i.e., perimeter monitoring. It is shown that the vehicle learns how to drive itself by simultaneously observing its surroundings and the actions taken by a human driver for a given task. That is achieved by embedding knowledge regarding perception-action couplings in dynamic representational states used to produce action flows. Thereby, representations link sensory data to control signals. In particular, the representational states associate visual features to stable action concepts such as turning or going straight.

Original languageEnglish
Title of host publication2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
ISBN (Electronic)978-1-5386-2939-0
ISBN (Print)978-1-5386-2940-6
DOIs
Publication statusPublished - 20 Oct 2017
Event14th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS 2017) - Carlo V Castle, Lecce, Italy
Duration: 29 Aug 20171 Sep 2017
Conference number: 14
http://www.avss2017.org/

Conference

Conference14th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS 2017)
Abbreviated titleAVSS 2017
CountryItaly
CityLecce
Period29/08/171/09/17
Internet address

Fingerprint Dive into the research topics of 'Dynamic representations for autonomous driving'. Together they form a unique fingerprint.

  • Cite this

    Olier Jauregui, J. S., Marin, P., Martin, D., Marcenaro, L., Barakova, E. I., Rauterberg, G. W. M., & Regazzoni, C. (2017). Dynamic representations for autonomous driving. In 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017 [8078511] Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/AVSS.2017.8078511