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

Area of Science:

• Automotive Engineering
• Network Architecture
• Embedded Systems

Background:

• Modern vehicles increasingly integrate advanced driver assistance systems (ADAS) and infotainment, driving demand for more electronic control units (ECUs).
• The proliferation of ECUs necessitates complex and heavy wiring harnesses for power and communication.
• Current domain-based in-vehicle network (IVN) architectures face challenges in managing this complexity for autonomous driving.

Purpose of the Study:

• To propose novel domain-based IVN architecture (DIA) and zone-based IVN architecture (ZIA) designs tailored for autonomous driving.
• To evaluate the effectiveness of ZIA in mitigating wiring harness complexity and data transmission delays within time-sensitive networking (TSN) contexts.
• To compare the performance of the proposed DIA and ZIA against existing solutions.

Main Methods:

• Development of new DIA and ZIA models for autonomous driving applications.
• Simulation of the proposed IVN architectures using the OMNeT++ network simulator.
• Analysis of wiring harness length/weight and data transmission delay metrics in a simulated mid-size vehicle environment.

Main Results:

• The zone-based IVN architecture (ZIA) demonstrated a reduction in wiring harness length and weight by approximately 24.6% compared to domain-based architectures.
• ZIA also resulted in a decrease in data transmission delay.
• Simulations were conducted within a realistic mid-size vehicle context.

Conclusions:

• ZIA offers a more efficient solution for managing in-vehicle networks in the era of autonomous driving.
• The proposed ZIA effectively addresses the challenges of wiring harness complexity and data transmission latency.
• This research provides a viable architectural approach for future automotive network designs.