Co-simulated digital twin on the network edge: A vehicle platoon

This paper presents an approach to create high-fidelity models suited for digital twin application of distributed multi-agent cyber–physical systems (CPSs) exploiting the combination of simulation units through co-simulation. This approach allows for managing the complexity of cyber–physical systems by decomposing them into multiple intertwined components tailored to specific domains. The native modular design simplifies the building, testing, prototyping, and extending CPSs compared to monolithic simulator approaches. A system of platoon of vehicles is used as a case study to show the advantages achieved with the proposed approach. Multiple components model the physical dynamics, the communication network and protocol, as well as different control software and external environmental situations. The model of the platooning system is used to compare the performance of Vehicle-to-Vehicle communication against a centralized multi-access edge computing paradigm. Moreover, exploiting the detailed model of vehicle dynamics, different road surface conditions are considered to evaluate the performance of the platooning system. Finally, taking advantage of the co-simulation approach, a solution to drive a platoon in critical road conditions has been proposed. The paper shows how co-simulation and design space exploration can be used for parameter calibration and the design of countermeasures to unsafe situations.