Research Highlights

The initial step involved developing C++/Python code to perform the forage approach operation. This code was first tested in simulation, allowing rapid and cost-effective validation of multiple variations without requiring physical access to the robot. As a result, the software was robust and largely validated before deployment. However, simulations cannot fully replicate real-world complexities.

Following successful simulation tests, autonomous navigation was implemented, enabling the robot to move independently in mapped environments—though it cannot yet operate in unknown areas.

Once navigation was confirmed, focus shifted to designing and integrating the blade. The first prototype, made of aluminum by a company in Brescia, underwent lab testing before being deployed at Baroncina Farm (Lodi) for field validation.

This motorized blade is controlled directly from the navigation PC via a manufacturer-provided driver. Initial issues included limited tilt and insufficient length. While the tilt was improved through design modifications, the decision was made to postpone length adjustments until confirming the tool’s suitability for forage handling.

For further details: fondazionelgh.it/robot-foraggiamento

Accurate modeling and simulation are essential for the design and validation of autonomous navigation algorithms. However, in agricultural applications, this task is particularly challenging due to the need for precise representation of tyre–ground or track–ground interactions.

This work aims to develop high-fidelity robot models, including detailed terrain representations, using an object-oriented, multi-physics modeling language such as Modelica. In parallel, it explores the development of computationally efficient yet accurate models using physics-informed neural networks.

Finally, the integration of models from both approaches with autonomous navigation systems—whether implemented in pure C++ or within a ROS2 framework—is also addressed.

For further details refer to the paper “Object-oriented modelling of a tracked vehicle for agricultural applications“.

Rice is one of the world’s most important staple crops, and in Europe, Italy is the leading producer, particularly in the northeastern regions. Traditionally, rice cultivation involves flooding fields from before planting until just before harvest. Maintaining this water level requires significant labor, as farmers must manually operate inlet and outlet gates. Moreover, this method results in high water consumption.

To address these challenges, new technologies using remotely and automatically controlled gates are being explored to improve irrigation efficiency. This research investigates the feasibility of an intelligent, synchronized gate system aimed at optimizing irrigation management and regulating ponding water levels more effectively.

For further details refer to the paper “The potential of a coordinated system of gates for flood irrigation management in paddy rice farm“.