Research On The Key Technologies Of VANET Positioning In City GNSS Blind Zones

VANET (Vehicular Ad-hoc Network) is the application of MANET (Mobile Ad-hoc Network) in traffic. Vehicles in the VANET can communicate with each other or communicate with roadside units through Short Range Communications. Thus, speed, position and other sensor data can be exchanged between the vehicles in VANET which ensures driving safety and improves road traffic efficiency. Since the information of location, trajectory and the relative distance between vehicles plays an important role in security-type applications of VANET. This dissertation aims to improve the accuracy of vehicle positioning and focuses on the key technology of the fast and accurate positioning of the vehicles in GNSS blind zones.GNSS blind zones means the areas where cannot receives accurate satellite positioning signals for the block of high buildings or tunnels. Vehicles usually obtain the location information with the help of GNSS. When a vehicle drives into the GNSS blind zone, it would lose its position, so INS (Inertial Navigation System) is always used to assist in positioning. As we all know, the error of INS would be bigger and bigger over time, finally makes the result with no effect. There are various problems of existing technology such as INS. roadside access points and assisted positioning etc. This article puts forward a collaboration positioning scheme of vehicles when in city GNSS blind zones. This scheme uses VANET as the architecture of vehicle communication. Doppler shift technology is adopted to measure the relative distance between the target vehicle and the other vehicles around. Then the qualified information was screened by some rules. Through these data we get several relative positions and finally get a rough estimate position by combining the results.The vehicle model and screening process adopt GRW model. GRW model presents GNSS state information of the vehicles and also provides restrictions of the information exchange. So that the exchange of information will be more simple and efficient. To some extent, GRW model improves the efficiency and effectiveness of the information calculation. The results would be calculated and filtered with unscented Kalman filter, which makes the results close to the real measurement situation. Then we combine the UKF results and the INS results to get the estimated position. Roadside positioning units is used in the areas with few vehicles or with large blind zone. They act as stationary high-precision positioning vehicles to make up for less GNSS signal source.The simulation results show that the proposed scheme can indeed improve the positioning accuracy of the vehicles while vehicles pass through the city GNSS blind zones.