Mapping and localization are both essential in the development of mobility applications for people or intelligent vehicles. These tasks that have been intensively studied in the past decade are considered among the most complex perception problems. They require a huge computation capability with strict real time contrains and also huge amount of data if large spaces have to be treated. This application also needs many actors and the delivered information must be valid and accurate in time and space, which implies frequent updates.
pLaTINUM project aims at developing a geolocalized global frame, stored on a cloud and made of a set of 3D representations based on geometric, photometric and semantic information. This geographical information system (SIG) must be able both to update and to enrich automatically its content thanks to data transmitted by remote agents and also to provide a navigation service. This double objective in this collaborative application consists in using information stored in the cloud in order to help the agents during their navigation tasks, which in return will inform the cloud about the local detected differences which will be treated afterwards in order to update the geolocalized central reference.
With this aim based on a collaborative system made of communicative agents able to at least perceive their environment such as smartphones, embedded cameras in vehicles or robots or even glasses, pLaTINUM prject is perfectly in adequation with the axis "Interaction humain-machine, objets connectés, contenus numériques, données massives et connaissances" of Défi 7 « Société de l’Information et de la Communication » for the following reasons :
- capture, eventually with multi-sensors, of the world by a network of intelligent and communicating agents in order to develop its numeric representation,
- the fusion of the data sent by communicating agents requires the collaborative aggregation of the different local representations into a global one, that must monitor the update and the enrichment of the content,
- the dynamic monitoring of the global representation requires the distributed capability of each agent to accurately localize itself, to detect significant changes of the environment, to mark up automatically significant elements and to transmit the results to the cloud,
- the interaction with this tool will the basis of new services dedicated to the mobility for people and intelligent vehicles.
Monsieur Pascal Vasseur (LABORATOIRE D'INFORMATIQUE, DE TRAITEMENT DE L'INFORMATION ET DES SYSTÈMES)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
Inria Institut National de Recherche en Informatique et en Automatique
LE2I Laboratoire Electronique, Informatique et Image
IGN Institut National de l'Information Géographique et Forestière
LITIS LABORATOIRE D'INFORMATIQUE, DE TRAITEMENT DE L'INFORMATION ET DES SYSTÈMES
Help of the ANR 732,931 euros
Beginning and duration of the scientific project: September 2015 - 42 Months