Traffic State Estimation with Bayesian Networks at Extremely Low V2X Penetration Rates

Abstract

In this paper the concept of Bayesian Networks (BN) is applied to the problem of traffic data acquisition by data fusion. Two wireless communication based sensors are used as data sources: IEEE 802.15.1 Bluetooth and IEEE 802.11p V2X (vehicle to vehicle and vehicle to infrastructure). Via V2X, so called cooperative awareness messages (CAM) are being received, which provide information on vehicle location and speed. Via Bluetooth, only the presence of a Bluetooth equipped device can be detected. Nowadays and in the near future, only a low number of road users is expected to be equipped with V2X. Therefore the rate of vehicles equipped is very low (around 1%). The equipment rate of Bluetooth devices is much higher. We assume, that between 5% and 50% of all road users can be detected ad reidentified with a Bluetooth scanning device. Bluetooth detectors have been notably used for traffic management purposed for years, e.g. for obtaining accurate journey times, but they have not been applied for Speed estimation so far. The approach of this paper is providing vehicle speed and vehicle count data by fusing moderate Penetration Bluetooth data and low penetration V2X data. The challenging task is to obtain accurate speed estimation data. Applying BNs for this purpose, we will show that the robustness of this stochastic fusion engine is capable of reaching speed RMSEs from 2 to 5m/s and complete the state estimation by 35% by fusing 1% V2X with 30% Bluetooth. The investigations are made on the basis of simulation.