Hybrid-ProbSense.KOM: Probabilistic Sensing with Hybrid Communication for Gathering Vehicular Sensed Data

Abstract:
Knowledge about vehicles’ surroundings gained by the use of local sensors is getting increasing importance. It provides the basis for many advanced driver assistance systems, that are constantly enhanced. However, the functionality of such systems is limited with the sensing range. In order to improve vehicular perception, communication can provide the means. Thereby, a long perception range can be provided by a backend that gathers vehicular sensed information via cellular communication. Such an approach also copes with sparse traffic situations. But since cellular communication is costly, transmission should be minimized. A centrally controlled probabilistic transmission mechanism, as introduced in [1], is able to reduce the amount of data traffic and simultaneously stick to certain quality metrics. Based on this we developed an extended approach, named Hybrid-ProbSense.KOM, that is based on cellular and intervehicle communication. By the additional use of inter-vehicle communication, we are able to show a further reduction in the overall cellular data traffic of up to 20%.