Cooperation is an enabler for autonomous vehicles. A promising application of cooperative driving is high-density platooning, where trucks drive with low inter-vehicle distances. It aims at increasing the road and fuel efficiency whilst guaranteeing safety. The safe and efficient coordination of the control requires the regular and reliable exchange of V2V messages. The performance of the vehicular application has been shown to be strongly affected by the variation of the performances of the communications system. To be able to adapt their functional settings to these variations, vehicles need the ability to predict it. We present a prediction model for the packet inter-reception time platoon messages in an IEEE 802.11p network. This performance indicator is the subject of extensive research as it captures the irregularity of input for the control loop. The prediction model uses conditional density estimation based on the exponential distribution. We fit this model using a multi-layer perceptron regressor based on features representing the surrounding communication environment. The presented results are based on data collected during a full scale platooning simulations using ns-3 and SUMO. We compare different environment abstraction models and show the potential of on-line learning. INDEX TERMS Vehicular ad hoc networks, intelligent vehicles, prediction algorithms, cooperative systems. GUILLAUME JORNOD received the B.Sc. and M.Sc. degrees in environmental science and engineering (ing. env. dipl. EPF) from the École Polytechnique Fédérale de Lausanne (EPFL), Switzerland, in 2012 and 2014, respectively, and the M.Sc. degree in management from University College London, U.K., in 2015. He is currently pursuing the Ph.D. degree in the field of vehicle-to-vehicle communications in conjunction with Technische Universität Braunschweig, under the supervision of Prof. Dr.-Ing. T. Kürner.

A recent feature of communications systems is the agile quality of service adaptation, in which the application and the communications system exchange requirements and prediction of quality of service. The application first provides its quality of service requirement. The communications system tries to enforce it, and makes a prediction of the available quality of service. Finally, the application adapts its settings to the future quality of service and provides updated requirements. Though this concept is originally designed for cellular-based technologies, it is also applicable to ad-hoc communication systems.In this paper, we focus on the prediction of quality of service for ad-hoc communications in a high-density platooning system. The quality of service of interest is the packet inter-reception time in an IEEE 802.11p network. Our platooning system drives through different vehicular traffic conditions, in which we gather transmission and position data. We then analyze the distribution of the packet inter-reception time to select the model features and then fit multiple distribution models. This empirical prediction modeling will then be the baseline for future modeling.

We present a benchmarking framework for different radio access technologies (RATs) in a high density platooning (HDPL) emergency braking use case. We assess the performance of IEEE 802.11p as well as LTE-V managed mode (mode 3) and unmanaged mode (mode 4) for sidelink communications. The performances are studied in terms of delays, packet error rates (PERs) and functional safety indicators. We first vary the number of vehicles, the surrounding traffic and the inter-vehicle distance. Multiple traffic scenarios are then investigated for the most challenging conditions. We find that for reasonable surrounding traffic, the platoon is generally safe in this emergency scenario, although packet error rates are growing for IEEE 802.11p and LTE-V mode 4 as the traffic intensifies, along with delays for the former technology. Thanks to scheduling, LTE-V mode 3 is not affected by this increasing PER and shows a large constant delay: the scheduling delay. With this study, we pave the way for a further study of these radio technologies with more accurate channel models as well as including new 5G components in our benchmarking.

No abstract

No abstract