Deterministic access to mobile edge for robot control

In modern industrial environments, the need for automated solutions increased as mobile network technology progressed. With the deployment of the fifth-generation standards (i.e., 5G), it is now possible to achieve a level of automation never reached before. 5G allows low latency, high throughput, and large bandwidth, which make it possible to deploy time-sensitive applications like autonomous driving and automated control of robots in an industrial environment. This thesis focuses on the deployment of a low-cost 5G network exploiting the tools provided by the open-source software OpenAirInterface (OAI) and the successive development of a Robot Operating System (ROS) 2 mobile robot application that could be deployed across the 5G network to automatically control the robot’s movement. Then, the performances of the system were studied, and the main criticalities were analyzed. Performance metrics, particularly latency, were analyzed as a critical factor in an automated application to grant the safety of the robot itself and of the human workers. Different implementations of the robot application were tested, and the best possible solution was proposed for the use case under investigation. The results demonstrate that reliable robot control can be achieved over a 5G network, meeting low-latency requirements. Future research could explore more agile robots, complex control laws, and commercial 5G systems with enhanced channel conditions and network slicing for increased reliability.