End-to-end driving via conditional imitation learning
Codevilla Moraes, Felipe (Centre de Visió per Computador (Bellaterra, Catalunya))
Miiller, Matthias (King Abdullah University of Science and Technology)
López Peña, Antonio M. (Universitat Autònoma de Barcelona. Departament de Ciències de la Computació)
Koltun, Vladlen (Intel Labs)
Dosovitskiy, Alexey (Intel Labs)

Imprint:	Institute of Electrical and Electronics Engineers (IEEE), cop.2018
Description:	8 pàg.
Abstract:	Deep networks trained on demonstrations of human driving have learned to follow roads and avoid obstacles. However, driving policies trained via imitation learning cannot be controlled at test time. A vehicle trained end-to-end to imitate an expert cannot be guided to take a specific turn at an upcoming intersection. This limits the utility of such systems. We propose to condition imitation learning on high-level command input. At test time, the learned driving policy functions as a chauffeur that handles sensorimotor coordination but continues to respond to navigational commands. We evaluate different architectures for conditional imitation learning in vision-based driving. We conduct experiments in realistic three-dimensional simulations of urban driving and on a 1/5 scale robotic truck that is trained to drive in a residential area. Both systems drive based on visual input yet remain responsive to high-level navigational commands.
Grants:	Agencia Estatal de Investigación TIN2017-88709-R Agència de Gestió d'Ajuts Universitaris i de Recerca 2017/FI-B1-00162
Rights:	Tots els drets reservats.
Language:	Anglès
Document:	Capítol de llibre ; recerca ; Versió acceptada per publicar
Subject:	Robot sensing systems ; Task analysis ; Vehicles ; Cameras ; Roads ; Navigation
Published in:	2018 IEEE International Conference on Robotics and Automation (ICRA), 2018, p. 4693-4700, ISBN 978-1-5386-3081-5

DOI: 10.1109/ICRA.2018.8460487

Postprint 9 p, 5.3 MB

The record appears in these collections:
Books and collections > Book chapters