Performance Evaluation of Zone-Based In-Vehicle Network Architecture for Autonomous Vehicles.

In recent years, various functions such as advanced driver assistance systems (ADAS) and infotainment systems are being mounted in vehicles for safety and convenience to drivers. Among the various functions, autonomous driving-related technologies are being added to all vehicles, from low options to high options. For autonomous driving, hundreds of new electronic control units (ECUs) including various advanced sensors would be needed. Adding more ECUs would enhance safety and convenience for the driver. On the other hand, wiring between these ECUs would be more complex and heavier. The wiring harness is essential for communication and power supply. Currently, the in-vehicle network (IVN) uses the domain-based IVN architecture (DIA) that separates ECUs into domains based on their functions. Recently, in order to minimize the complexity of wiring harness and IVN, zone-based IVN architecture (ZIA) that groups ECUs according to their physical locations is attracting attention. In this paper, we propose a new DIA and ZIA for autonomous driving in the context of time-sensitive networking (TSN). These two new IVN architectures are simulated using the OMNeT++ network simulator. In the simulation process, a mid-size vehicle is assumed. It is shown in this paper that ZIA not only reduces wiring harnesses in both lengths and weights by approximately 24.6% compared to the DIAs, but also reduces data transmission delay.