logo EDITE Sujets de doctorat

Internet-of-Vehicles for Connected Autononomous Driving

Sujet proposé par
Directeur de thèse:
Encadré par
Doctorant: Raj Haresh PATEL
Unité de recherche UMR 7102 Laboratoire de recherche d'EURECOM

Domaine: Sciences et technologies de l'information et de la communication

Projet

Autonomous vehicles have been presented during the GM Futurama in 1939, and their technology specifications formally designed already in the 80’. Such effort could unfortunately not reach production state mostly due to the cost of the required communication (analogic radio) and detection radars. The deal partially changed with the appearance of affordable positioning (GPS), communication (DSRC), and data warehouses from the Internet Cloud. As demonstrated by Google and its driverless car, or more recently by major car manufacturers (BMW F+T, Daimler, Tesla...), control mechanisms driving a vehicle autonomously are nowadays mature. But a long road remains ahead of us before we can see autonomous vehicles in the streets. Beside legal and societal aspects, the major reason still comes from the significant cost of providing the data required by control mechanisms to safely navigate vehicles (e.g. the cost of Google's rotary lasers (LIDAR) is approx. 80’000 euros).

Recent appearance of dedicated short range communication (DSRC) between vehicles, using a WLAN extension to vehicular environments (IEEE 802.11p), showed to be a flexible method to interconnect vehicles to exchange data, for instance to improve traffic safety or provide smart mobility solutions. Through DSRC and in cooperation with cellular technologies, vehicles may not only connect to Internet, but be part of an Internet-of-Vehicles (IoV).

From a communication perspective, the IoV opens doors to large amount of data available in the cloud or between vehicles that could be available to connected autonomous driving (CAD) vehicles. From a mobility perspective, cooperative mobility strategy could be devised to enhance the performance of autonomous navigation.

In the context of this thesis, we propose to analyze the benefits of an Internet-of-Vehicle built from DSRC and 4G/5G networks to provide precise, secured and affordable data required by the control mechanisms of CAD vehicles.

Enjeux

During this Ph.D, we will investigate and provide technical solutions to the problem of scalable and reliable data in the IoV for CAD vehicles.

In particular, we will:

  • Propose an evaluation methodology to identify the requirements of autonomous vehicles, and how cooperative data from the IoV could (partially) replace autonomous detectors (radars, lidars).
  • Evaluate the feasibility of DSRC to provide such data itself, and by cooperating with 4G/5G networks for accessing the cloud.
  • Prototype the proposed solution on the SDR platform OpenAirInterface (www.openairinterface.org/‎)
  • Ouverture à l'international

    EURECOM has international cooperation through its academical and industrial partners, and mobility between partners is encouraged.

    Remarques additionnelles

    Thématique: H