Thesis
The recent advent of connected vehicle platforms permits the distribution of infotainment assets to drivers and passengers with pulled and pushed workflows in a comparable manner to current mobile handsets. However, vehicles differ technically from mobile phones in terms of capability and in terms of usage.
This thesis tackles the subject of personalised media delivery to motorists. We first study the technical characteristics of vehicle infotainment platforms, media assets and metadata in order to identify the requirements of a media delivery architecture for a vehicle. Based on those constraints, we have specified a media on-demand framework, which has been developed in a prototype.
Afterwards, we tackle the topic of personalisation in light of two complementary point of views: On the one hand, the driver can process active personalisation when using a proper human machine interface. We present a music browser for online libraries that allows the creation of multicriteria playlists while driving. On the other hand, we analyse passive personalisation, which makes use of the driving context. We discuss the repartition of the functional components and build up a distributed architecture, which takes into account individual context preferences and their integration in the multimedia architecture that we have formerly presented. Eventually, the different solutions are evaluated according to experimental and expert methods.
Type:
Thèse
Date:
2011-05-30
Department:
Data Science
Eurecom Ref:
3406
Copyright:
© TELECOM ParisTech. Personal use of this material is permitted. The definitive version of this paper was published in Thesis and is available at :
See also:
