Performance Evaluation of Vehicle Positioning System using Multi-sensor in GNSS Blockage Area

摘要

Vehicle navigation systems, which rely on GNSS(Global Navigation Satellite System) signal, have limitations that it is impossible to position the vehicle in GNSS signal shaded areas such as building forests and tunnels. Due to the development of MEMS(Micro-electromechanical Systems) technology, a variety of low-cost MEMS-IMU(Inertial Measurement Unit)s have been launched, and studies combining GNSS and MEMS-IMU have been actively conducted. MEMS-based IMU has an advantage in terms of the cost, but its error rapidly increases in a short time, compared with the conventional inertial sensor. To overcome this problem, additional sensors are needed to obtain more observations(Brown, 2005). Recently, vehicles equipped with ADAS(Advanced Driver Assistance System) have been released for the convenience and the safety of the driver(Heo, 2009). To run the ADAS, vehicles with various sensors, such as wheel speed sensors, yaw rate sensor and steering angle sensors, are on the market. Therefore, it is expected that the positioning performance will be improved if additional observations are utilized by using various sensors as described above. In this study, GNSS, MEMS-IMU, on-board vehicle sensor based car positioning system is developed. The vehicle positioning algorithm is implemented based on an extended Kalman filter and adopts a closed loop. As the result, compared with the GNSS and MEMS-IMU combination, performance improves greatly depending on whether the wheel speed sensor is added in the GNSS signal blockage. The most stable positioning performance was confirmed when all available sensors were used. In the 45seconds of 10 GNSS signal blockages, mean of horizontal error was 5.31m.