Show simple item record

dc.contributor.authorMüller, Matthias
dc.contributor.authorDosovitskiy, Alexey
dc.contributor.authorGhanem, Bernard
dc.contributor.authorKoltun, Vladen
dc.date.accessioned2019-04-28T13:12:39Z
dc.date.available2019-04-28T13:12:39Z
dc.date.issued2018-04-25
dc.identifier.urihttp://hdl.handle.net/10754/632519
dc.description.abstractEnd-to-end approaches to autonomous driving have high sample complexity and are difficult to scale to realistic urban driving. Simulation can help end-to-end driving systems by providing a cheap, safe, and diverse training environment. Yet training driving policies in simulation brings up the problem of transferring such policies to the real world. We present an approach to transferring driving policies from simulation to reality via modularity and abstraction. Our approach is inspired by classic driving systems and aims to combine the benefits of modular architectures and end-to-end deep learning approaches. The key idea is to encapsulate the driving policy such that it is not directly exposed to raw perceptual input or low-level vehicle dynamics. We evaluate the presented approach in simulated urban environments and in the real world. In particular, we transfer a driving policy trained in simulation to a 1/5-scale robotic truck that is deployed in a variety of conditions, with no finetuning, on two continents. The supplementary video can be viewed at https://youtu.be/BrMDJqI6H5U
dc.publisherarXiv
dc.relation.urlhttp://arxiv.org/abs/1804.09364v1
dc.relation.urlhttp://arxiv.org/pdf/1804.09364v1
dc.rightsArchived with thanks to arXiv
dc.titleDriving Policy Transfer via Modularity and Abstraction
dc.typePreprint
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.contributor.departmentElectrical Engineering Program
dc.contributor.departmentVisual Computing Center (VCC)
dc.eprint.versionPre-print
dc.contributor.institutionIntelligent Systems Lab, Intel Labs, Germany
dc.contributor.institutionIntelligent Systems Lab, Intel Labs, USA
dc.identifier.arxivid1804.09364
kaust.personMüller, Matthias
kaust.personGhanem, Bernard
refterms.dateFOA2019-04-29T06:50:28Z


Files in this item

Thumbnail
Name:
1804.09364v1.pdf
Size:
4.429Mb
Format:
PDF
Description:
Preprint

This item appears in the following Collection(s)

Show simple item record