Co-simulation of an autonomous driving rover for agro-industrial applications.

The objectives of this thesis are multiple and well defined, aimed at solving the problems related to the modeling, simulation, and control of an autonomous differential traction rover. First, the work aims to develop a detailed kinematic model of the rover using Simulink. This model will be based on mathematical equations that describe the movement of the vehicle, considering its differential traction configuration. Creating a robust kinematic model is necessary understanding the basic dynamics of the rover and will form the basis for subsequent developments in control and dynamic simulation. Another important objective of the thesis is to implement path following in Simulink. These algorithms will be crucial to allowing the rover to follow predefined paths with precision. The approach used for the algorithm is point-2-point and the implementation will require the development of a control system capable of adapting to various operational scenarios, maintaining the stability and precision necessary to navigate through a series of waypoints. This part of the work is critical to ensuring that the rover can operate autonomously and precisely, responding effectively to programmed route inputs. The dynamic modeling of the rover with Adams view constitutes another key objective of the thesis. This model will include detailed mechanical and physical components of the rover and of the environment, such as tire-ground interaction. Using Adams for this modeling permits to obtain more accurately simulate the behaviour of the rover under realistic operating conditions, considering interactions with terrain of every kind. This phase is essential to obtain a realistic representation of the rover’s performance in complex environments. The integration between Simulink and Adams to create a co-simulation platform represents a further crucial objective. This integration will allow a continuous exchange of data between the control model in Simulink and the dynamic model in Adams, improving the accuracy and realism of the simulation. Co-simulation will allow analyzing the impact of control strategies on the dynamic behaviour of the rover, providing a complete view of its performance, and facilitating the optimization of the overall system. The validation of the integrated model and the analysis of the simulation results constitute another important objective of the thesis. This phase consists of the comparison of the results obtained with kinematic model respect to results obtained with the model integrated to multibody system and verify the precision and the reliability of the system. Analyzing the results is crucial to identify any areas for improvement and further optimize the rover’s control system and dynamics. Another objective is the application of the co-simulated model in virtual environments created with Adams. This will allow the rover to be tested in complex operational scenarios, such as agriculture fields with irregular terrain and variable obstacles. Demonstrating that the model can be used as a practical tool for the development and optimization of autonomous vehicles in real applications is the goal of the thesis.