Optimized Autonomous Navigation for Field Robots: Extended Results and Practical Deployment

Abstract This study introduces an optimized algorithm for the autonomous navigation of field robots, designed to deliver faster and more accurate performance in agricultural environments. The primary goal was to advance a previously developed algorithm by improving navigation accuracy, reducing crop damage, and shortening execution times. The proposed solution integrates advanced data filtering methods with sensor fusion, combining LiDAR and IMU inputs to generate precise 3D point cloud representations for reliable navigation in structured crop rows. Both the legacy and the new algorithms were evaluated through simulation and real-world trials on the FarmBeast robotic platform. Experimental results show that the improved algorithm reduces traversal time by up to 33% on certain field sections and lowers crop damage by 25%. These findings confirm the robustness and effectiveness of the system in handling complex agricultural field conditions, underscoring its potential for practical use in farming automation.