GPS Positon Predicting System by Kalman Filter with Velocity from OBD and Direction from Magnetometer

Abstract

In this research, the GPS position system improving in a good and bad environment with Kalman filter is presented. Which proposed system consist of a GPS L26-DR module, GY-26 compass sensor for detected direction, and request velocity data from ECU with OBD-II and used as an input to Kalman filter to predict a more accurate position. The proposed system operates with two modes, the first mode is the mode that received normally position data, the received position data is fed to be an input of the Kalman system to improve that position data. If it loses the GPS position, the last position received from GPS will be used as an input to Kalman filter to predict the next position and after that, the previous predicted position data is feedback as an input to Kalman filter, and it repeats the process until the system can receive good GPS signals once again. The second one operates as the first mode in a normal situation, but the difference is, when it loses the GPS position, velocity data from a vehicle, direction from compass module and the last GPS position will be used to predict the next position. When it gets normally GPS position data, the GPS position is fed to an input of the Kalman filter. The results show that the first model has more errors than the second one but it's better than only conventional GPS positioning.