Designing a far-reaching view for highway traffic scenarios with 5G-based intelligent infrastructure

Gereon Michael Hinz, Martin Buechel, Frederik Diehl, Guang Chen, Annkathrin Krämmer, Juri Kuhn, Venkatnarayanan Lakshminarasimhan, Malte Schellmann, Uwe Baumgarten and Alois Knoll

8. Tagung Fahrerassistenzsysteme,

November 2017 · Munich, Germany

abstract

Cooperative vehicle infrastructure systems offer significant potential for improved traffic safety, throughput and improved energy efficiency. Infrastructure sensors along the road can substitute vehicular sensor-sets, providing improved robustness and performance through different mounting positions and orientations, reducing occlusions, stationary locations, facilitating system-wide calibration, optimization for the specific traffic area in view, and vastly increase perception range by combining multiple measurement points. Communication via fifth Generation (5G) networks offers solutions to the corresponding substantial requirements for high bandwidth, low latency and high reliability for data and information communication. We propose a concept, which aims to provide a far-reaching view to (self-driving) vehicles and drivers with infrastructure sensors and 5G communication, as a cognitive system. The system detects and localizes traffic objects and predicts their future movements. The resulting information will be provided to traffic participants allowing for safer, more proactive and comfortable driving.

subject terms: V2X, autonomous driving, connected cars, 5G, sensor-sets, far-reaching view, big data, environmental perception

url: https://mediatum.ub.tum.de/doc/1421303/1421303.pdf