Using extended Kalman filtering (EKF) to fuse GPS and inertial navigation systems (INS) for land vehicle control.

摘要

When Global Positioning System (GPS) is used in land vehicle applications, the so-called GPS latency delays the output of the vehicle position data. An algorithm based on extended Kalman filtering (EKF) is added to an existing GPS-based automatic steering system for agricultural tractors to reduce the effect of the GPS latency. This work is built upon the analytic framework developed earlier by Bouvet and Gacia.; The question to be answered is: Can extended Kalman filtering reduce the latency and improve the system performance? In this thesis, the existence of the GPS latency is taken into consideration in the system modeling, the GPS latency is measured, and the vehicle motion, the GPS observation, and the vehicle turning models are established and simulated. From the results of the simulation and analyses, it is concluded that the EKF can reduce the effect of GPS latency and improve the system performance. Although the EKF showed some improvement as compared to the current control performance, its practical value was deemed insignificant.