5G networks are encountering virtualization technologies as the foundations of the softwarization of the infrastructure. The usage of these techniques in the Connected and Automated Mobility (CAM) vertical is the key to address mobility and computing issues. The next generation of CAM services are demanding continuous sensor-data gathering and processing, but current solutions lack of flexibility and computing capabilities in the On-Board Units (OBUs). Consequently, a dynamic intermediate stratum with adaptable networking resources and data processing offloading is required to cover the requirements imposed by the upcoming vehicular applications and users. Besides, due to the changing nature of these environments, dynamic testing and validation of the deployed services is necessary to assure their correct functioning. In this line, a solution that exploits the Multiaccess Edge Computing (MEC) paradigm to instantiate virtual OBUs (vOBUs) to act as virtual counterparts of the physical ones is presented. By doing so, in-vehicle OBUs can be protected from the characteristic disconnections of vehicular networks using the vOBU as an intermediate communication layer. Besides, they can offload heavy computing processes to the edge. The solution is dynamically deployed as a Network Application (NetApp) in a real 5G testbed in the context of the 5GASP project, in which it is also possible to test and evaluate the functioning of the NetApp after the deployment.

This work has been supported by Fundacion Seneca—Agencia de Ciencia y Tecnologıa de la Region de Murcia—under the FPI Grant 21429/FPI/20, and co-funded by Odin Solutions S.L., Region de Murcia (Spain); by the Spanish Ministry of Science and Innovation under the DIN2019-010827 Industrial PhD Grant, and co-funded by Odin Solutions S.L.; and by the European Commission under the 5GASP (Grant No. 101016448) project.