Path tracking control of paddy field weeder integrated with satellite and visual methods

Abstract

To address the issue of insufficient path tracking precision and the tendency to deviation from the pre-determined path in autonomous field weeders. A dynamic adaptive control algorithm employing a three-dimensional fuzzy controller has been developed. The particle swarm optimization support vector regression (PSO-SVR) algorithm was utilized to predicts the front wheel steering angle of the paddy field power chassis, Based on the prediction results, a steering angle feedback compensator controller was designed. Navigation control factors were constructed accordance with navigation deviation, and these factors were adopted as navigation switching conditions for an integrated navigation mode. A navigation simulation model incorporating the feedback compensation control algorithm was established, and a comparative simulation analysis between integrated navigation and visual navigation was conducted. The simulation results indicate that the navigation error and stability of the proposed method are superior to those of visual navigation at different speeds. The effectiveness of the proposed algorithm was validated through field experiments. The results of field experiments reveal that, in contrast to visual navigation, the proposed integrated navigation mode improves navigation accuracy at different speeds, with the standard deviation of lateral offset reduced by 11.3% to 40.4%, and the standard deviation of heading angle reduced by 5.9% to 16.8%. These results suggest that our approach not only ensures the stable tracking of the target path for the paddy field weeder but also effectively enhances the navigation accuracy.