Design and field validation of a Dual-Axle steering system for autonomous tractors

The adoption of autonomous robots is steadily increasing, offering innovative solutions to pressing agricultural challenges. However, most robots rely on electric power and differential steering, limiting their applicability in demanding environments. This work presents a diesel- and hydraulically-powered autonomous tractor specifically designed for fruit tree plantations, incorporating a mechanical steering system with dual steerable axles. In an environment operated in ROS, the embedded sensors feed data to the electronics that activate pumps and electro-hydraulic valves to perform operations. The steering system allows for three different modes: two commonly used, Front Steering (FS) and Front-Rear Inverse Steering (FRIS) and a new one proposed in this work to improve the performance of the two previous modes, Hybrid Steering mode (HS). To evaluate the modes the execution of the planned route was checked for a straight-line and for different turning radius varying the number of waypoints. After determining the best configuration for real conditions, the autonomous tractor was tested in an intensive olive orchard. Results from the obstacle-free outdoor tests showed that FRIS was the most suitable steering mode to complete different radius turns with higher accuracy (1.88 ± 0.44 m) and minimal manoeuvres and HS for straight-line (2.23 ± 0.41 m). The autonomous navigation test in a commercial plot showed low Fréchet distances (1.56 ± 0.10 m) with low operating times, demonstrating the feasibility of the given configuration. The results obtained can help establish the necessary control method for autonomous machinery of similar fruit trees.