Simulation of the dynamic behavior of a self-driving rover for off-road applications

Technological advancements in agriculture are continuously evolving to meet the growing demand for food. In recent years, agricultural technology has been shifting towards a new paradigm: Agriculture 4.0. However, a new role within this system needs to be introduced. Various rover models have been developed in recent years to enable autonomous decision-making. This thesis focuses on studying the dynamic behavior of such vehicles in an environment that simulates real-world conditions, proposing improvements. The first phase explores the steering systems commonly adopted in agriculture, followed by a brief theoretical explanation of their operation, evaluating which system might yield better results theoretically. Subsequently, multi-body simulations are conducted to study the dynamic behavior, reporting contacts and a structure as close as possible to the real rover. For this purpose, a simplified model of an agricultural rover was developed using multi-body software. The study specifically compares the behavior of three rovers employing three different steering systems. At the end of the analysis, the best system from a dynamic perspective will be defined by observing variations in forces, lateral deviation, and other parameters that will be introduced, while also considering the feasibility of the existing model and the economic cost required for implementation.