Control coordination and monitoring of autonomous agents in agri–food field

Modern autonomous and distributed systems play an increasingly important role in applications ranging from precision agriculture to smart transportation and industrial automation. Ensuring reliable coordination, security, and fault management in such systems is a significant challenge due to practical limitations including communication delays, quantization, measurement noise, and partial observability. This thesis develops theoretical frameworks and methods to address these challenges, with all results validated through comprehensive numerical examples. The thesis first focuses on the controlled and coordinated behaviour of multiple agents under realistic operational constraints. Robust controllers are designed to achieve consensus and stability in networks of nonlinear agents, such as fleets of Unmanned Aerial Vehicle (UAV)s for agricultural monitoring, while mitigating the effects of disturbances and measurement errors. The proposed approaches ensure that even under practical imperfections, agents can operate cohesively and effectively. It then examines security threats in Discrete Event Systems (DES)s, addressing both active attacks that disrupt system operation and passive attacks that attempt to infer sensitive information. Methods are developed to detect and localize these attacks, preserving system integrity in both single and networked–systems. This work provides formal conditions and practical strategies to maintain reliability despite potential malicious interventions. Finally the thesis investigates fault detection under partial observation, where sensor information is limited. Observer–based schemes are developed to identify input faults using limited sensor information, providing tools to enhance resilience and offering practical tools for monitoring complex systems. Overall, this thesis presents novel methods to control, monitor, and secure autonomous systems, with the goal of making them more reliable, safe, and resilient in real–world conditions. Although the work emphasizes applications in agriculture, the approaches are f lexible and can be applied to a wide range of autonomous technologies. The findings offer a strong theoretical foundation for building systems that can operate effectively and robustly in practical environments.