Global positioning of off-road vehicles by sensor fusion for precision agriculture

This research utilized two optical encoders as local sensors, and a low cost GPS receiver as a global sensor. The GPS and encoders were integrated to form a complementary sensor pair for high accuracy positioning. The software for sensor fusion was composed of two parts: (1) a dynamic model to integrate GPS and optical encoders, which is based on the vehiclepsilas kinematics; (2) a time varying Kalman filter to merge sensor data. Both simulations and experiments were carried out to investigate the effects of filtering noise. The results showed that the positioning system designed worked very well, could accurately estimate vehicle position and heading angle in real time. The noise mixed with the GPS signal was reduced by 80% with filtering for practical applications.