A vehicle cooperative positioning enhancement for GNSS-dark areas based on GRW model

This paper proposed a vehicle cooperative positioning enhancement for GNSS-dark areas. The vehicle periodically exchanges the measurement and status information with neighbor vehicles within one-hop-range by the Dedicated Short Range Communication (DSRC) devices. When the GPS/BDS signals are blocked or degraded greatly when a vehicle travels in downtown areas among tall skyscrapers or through tunnels, its navigation system restarts fusing the reference GPS/BDS with dead-reckoning system, where the reference GPS/BDS is determined according to the neighbor´s GNSS information provided by the periodically exchanging messages, and their relative distances as well as motion directions evaluated via Doppler frequency shift. GNSS rank weighting (GRW) model is introduced to distinguish the reliability of the estimated reference GPS/BDS and assist in choosing the proper ones. Discontinuity, vehicles cannot receive satellite positioning signal in urban road because of the high buildings and tunnels. This problem has influenced many vehicles. Usually, inertial navigation system is used, but the effects is not satisfactory. In this article, we construct vehicular ad-hoc network with vehicle nodes and use dedicated short range communications (DSRC) technology to integrate messages and screen data. GNSS rank weighting (GRW) model is proposed to distinguish vehicular state. Vehicular relative position and velocity can be estimated by Doppler frequency shift. Finally, unscented Kalman filter (UKF) is employed to optimize INS to obtain accurate and efficient location information. The simulation results show that our proposed positioning approach can improve the positioning accuracy to some extent when the GNSS of the target vehicle suffers outages and errors.