ABSTRACT

Connected and autonomous vehicles run their control algorithms in dedicated on-board computing platforms, which will become obsolete long before the end of the life cycle of the vehicles, severely limiting the evolution of their control software and the deployment of cooperative vehicular applications. A promising solution for this problem is to delegate the most demanding computational tasks to the edge nodes of the Vehicular Ad-hoc Networks, leveraging on the Vehicular Edge Computing paradigm. This requires both low-latency, high-bandwidth communications and secure computing offloading. In this paper, we propose an architecture that ensures supporting secure computation offloading, using the IOTA-VPKI security scheme, without additional delay overhead. Furthermore, to demonstrate the applicability of the proposed scheme to a real case, we measured the time required for the execution of a maneuver plan supervised by an application, the Maneuver Control (MC), located on an edge node. Experimental results show that the described scheme is a very promising solution.

KEYWORDS

Computation Offloading, Connected and Autonomous Vehicles, Edge Computing, Distributed Ledger Technology, Security