Experimental Characterization and Co-Simulation Analysis of an Agricultural Rover with a Fuel-Cell Range-Extender Unit

The adoption of autonomous-driving rovers represents a feasible solution to improve the sustainability of the agricultural sector, as they are smaller and more efficient compared to traditional machinery. However, endurance and productivity can be critical factors for the adoption of such vehicles. In addition, the autonomous-driving algorithm should guarantee that the rover is able to accomplish tasks without supervision. In this paper, a numerical analysis of an autonomous-driving rover with a hybrid fuel-cell powertrain, specifically designed for orchards and vineyards, is presented. The proposed powertrain presents a first innovative integration of a metal-hydride hydrogen-storage system into an orchard mobile machine. A Li-ion battery pack is the main energy source, while the fuel-cell system operates in a range-extender configuration. A co-simulation model was developed comprising the autonomous-driving algorithm, a multibody model, and a powertrain model. Experimental tests were carried out to characterize the fuel-cell system and the metal-hydride tank, and the obtained data were used to develop and tune their numerical models. A virtual test scenario consisting of a typical rover maneuver, namely a 180-degree turn, performed in different soil and payload conditions, was defined, and simulations were carried out evaluating the rover’s performance. The simulation results showed that the rover completed the mission in loam and hard soil conditions, and with up to 200 kg of payload. Moreover, the fuel-cell range extender significantly enhanced the rover’s endurance, with up to +60% of increase when employing a tank swap technique to replace the metal-hydride tank upon hydrogen depletion. On the contrary, in the case of critical terrain conditions, such as muddy and sandy soils, the rover was not capable of completing the task due to tire slipping.