Vehicle self-localization for advanced driver assistance systems

In this paper a vehicle self-localization approach in an urban traffic intersection scenario is proposed. A sensor data fusion of azimuth angle of arrival (AOA) and estimated distance measurements with speed and heading information using an Extended Kalman Filter (EKF) is carried out to track the vehicle location. Furthermore, a concept for cooperative localization (CL) is introduced based on device-to-device (D2D) communication with a spatial based selection algorithm, which classifies the significance of received signals from neighboring vehicles for cooperative localization. For simulation purposes an inter-cell interference coordination (ICIC) is deployed in the network planning to mitigate the effect of interference from neighbor sectors of the same base station and from cells of neighbor base stations. Simulation results show that the proposed vehicle self-localization approach provides a lane accurate localization of the vehicle within 150m radius with respect to the traffic intersection origin considering inter-cell interference and multipath propagation.